The current study shows that, for South American tropical forests, tree alpha diversity is also associated with geological age of substrate. While subsidiary to rainfall effects, alpha diversity tended to be higher on younger substrates, which in turn were more fertile substrates, even after accounting for precipitation variability (Figure 10.5d). This is almost certainly not an effect of age per se, but rather due to the correlation between age and fertility in Amazonian surface geology. This effect can be seen most clearly in Amazonas state of Venezuela and adjacent Brazil and Colombia, where forests on shield areas have much lower alpha diversity than those on adjacent Tertiary sediments, even though total precipitation and precipitation variability are similar. The result that the highest diversity forests in Amazonia are on some of the youngest substrates emphasizes the role of assembly through plant migration rather than in situ diversification (Wilf et al., 2003; Ricklefs, 2004) (though Richardson et al., 2001 present an interesting counterexample in the genus Inga). If tropical forests are museums of diversity, they are museums where the exhibits are constantly rearranged.
Substrate also has a large effect on the floristic composition of Amazonian tree communities, and these effects are conserved at higher phylogenetic levels. At a local spatial scale, at our upper-Amazonian sites, tree plots can be reliably classified to floodplain or terra firme forest even if stems are only identified to family, and that result holds generally at localities across the Amazon Basin (e.g., Terborgh et al., 1996). Terborgh and Andresen (1998) showed that at larger spatial scales, however, adjacent terra firme and floodplain sites were more similar to each other than either was to the same habitat type at a more distant site. Floodplain sediments by and large reflect local to regional sediment transport and suggest that trees are responding to soil characteristics. Another explanation would be that trees are highly dispersal-limited, and that floristic differences among regions reflect in situ evolutionary differences (Campbell, 1994). While biogeographic explanations cannot be completely discounted, collecting expeditions to areas of similar geologies, even areas separated by hundreds to thousands of kilometers, have surprisingly similar floras (Schulenberg and Awbrey, 1997; Neill, 1999; Holst, 2001; P. Nunez, unpublished). Conversely, juxtaposed areas of distinct geologies show distinct species compositions. At the southern limit of the Amazon Basin in Bolivia, rainforest in Madidi National Park shows more similarity to forests derived from Andean sediments ^2,000 km away in Ecuador than the Brazilian Shield forests at the same latitude in Noel Kempff Mercado National Park, a distance of ^500 km (Pitman et al., 2001; Macia and Svenning, 2005; Silman et al., 2006).
Soils and their underlying parent materials affect diversity in two ways. The first is in the total diversity at a point, or alpha diversity. Gentry (1988) presented empirical data that suggested that alpha diversity is highest on rich soils, though the difference in soil fertility was much less important than precipitation amount and seasonality. This result was borne out by Clinebell et al. (1995) and the results of the current study. A second way that soils and geology influence diversity is through their effects on species distributions through niche relations. In this case taxa have preferences for soil types, leading to different species composition among soil types, with this beta diversity increasing the total (gamma) diversity of a region. Thus, edaphic effects on diversity can act through ecological processes at the hectare scale, and through floristic (distributional) effects at larger spatial scales.
Changes in community composition and plot-to-plot similarity with geological substrate are well-known in both the Neotropics and Paleotropics (e.g., Duivenvoor-den, 1995; Clark et al., 1998; Potts et al., 2002; Phillips et al., 2003; Tuomisto et al., 2003; Palmiotto et al., 2004; Valencia et al., 2004; Masse, 2005; Russo et al., 2005). Tests of substrate effects will need to be carried out at multiple spatial scales because substrate geology can affect plots; both through species-level ecophysiological effects (the niche) and larger-scale and longer-term effects on local species pools. Physiological and related ecological niche effects are likely due to direct effects of nutrient status or water holding capacity on species ability to maintain a positive population growth rate on a certain substrate, or indirect effects of natural enemies mediated through physiological effects (e.g., Givnish, 1999; Fine et al., 2004).
The role of the dynamic pre-Quaternary geological history of the area encompassed by modern Amazonia in generating modern patterns of plant diversity is only recently being integrated into studies of floristics and diversity (Fine et al., 2005). However, phylogeographic explanations in light of the Andean orogeny and its associated effects on the continental margin and shield areas have become standard explanations of animal diversification in Amazonia. Much of the lower Amazon was under water until 2-5 Myr bp due to high Miocene and Pliocene sea levels, and western Amazonia was a sequence of depositional centers very different from today, with a mosaic of shallow lakes and seas (Potter, 1997; Kronberg et al., 1998; Rossetti et al., 2005). The rapid uplift of the Bolivian Andes, rising —3 km in elevation between -10.3 and 6.8 Myr bp (Lamb, 2004; Ghosh et al., 2006), and the dynamism of the Amazonian forelands throughout western Amazonia (Rasanen et al., 1992; Kronberg et al., 1998; Hovikoski et al., 2005; Roddaz et al., 2005) certainly had profound effects on Andean and Amazonian phytogeography, though they remain unexplored.
Was this article helpful?